The author examines the relationship between research productivity and teaching effectiveness using data from the Erasmus School of Economics. The initial findings indicate a positive overall relationship between the variables. A more detailed analysis reveals a sharp reversal in the nature of the relationship. Although the relationship is negative in the first two years of the bachelor's programs, it becomes positive later on. This suggests that the applicability of the various models that have been advanced to explain either a negative or positive relationship may depend on course level. The result also has implications with respect to the most effective allocation of faculty in bachelor's and master's programs.